GnuPG 2.0.27 fails tests on Solaris 10 sparc

[Laurent Blume]

On 19/02/2015 16:43, Werner Koch wrote:
>
> Did 2.0.26 build on the same machine?
>
> Maybe the Pth problem which has already been reported?

I think it's actually related to this libgpg-error issue, which is still 
present in 1.18:
http://lists.gnupg.org/pipermail/gnupg-devel/2014-December/029226.html

Because I checked that an old 2.0.26 binary, which had passed the tests 
previously, is now having the symptoms.

>
> Please run common/t-convert under a debugger and show the stack
> backtrace.
>

Can we have a look at this? Even if the problem is not the same, at this 
point, I'd rather get it out of the way if only to be sure.

The current core looks the same as what it was then:

$ file core
core:           ELF 32-bit MSB core file SPARC Version 1, from 't-printf'

$ pstack core
core 'core' of 29157:   ./t-printf
  ff25c7fc __lwp_sigqueue (6, ffffffef, ffffffec, ffbff690, 5, 6) + 8
  ff1db5ec abort    (ff3540fc, ff2f0234, 6, 1, ff2f1b48, 0) + 108
  ff3540c8 get_lock_object (ff36f4a4, ff353ac4, ff3578e0, ff3ee388, 1, 
c21) + 4
  ff354100 _gpgrt_lock_lock (ff36f4a4, 0, 50, ff322a40, ffffffff, 0) + 4
  ff3577d8 _gpgrt_fflush (0, ff36f4a4, ff2f4840, 0, 1, 0) + d4
  ff3578e0 do_deinit (ff2f6e40, 0, ff322a40, ff322a40, ff320140, 2944) + 10
  ff1dd00c _exithandle (ff320000, ff320100, ff2f4840, 2880, 2800, f) + 50
  ff1cb1ac exit     (0, 142e0, f, f, f, 15) + 4
  00010f10 _start   (0, 0, 0, 0, 0, 0) + 64


Thanks!

Laurent